Virtual Doctor

Team Member Are :
Ashwini Pai
Shawn A Lobo
Machaiah A
Joy A Sequiera

Project Guide: Dr. Demian Antony D’Mello Abstract: This project is an application of telemedicine to alleviate primary health care problems in around the world where medical assistance is scarce. The project named 'The Virtual Doctor Project' will use mobile devices or PDA’s fitted with satellite communication devices and modern medical equipment to deliver primary health care services to some of the neediest areas of the country. The relevance and importance of the project lies in the fact that these areas are hard- to-reach due to rugged natural terrain and have very limited telecommunications infrastructure. The lack of these and other basic services makes it difficult for medical personnel to settle in these areas, which leads to an acute shortage of medical personnel. Virtual doctor application is capable of providing real time medical assistance for those seeking medical help. This application is intended to provide accessibility in both online and offline mode. It will be connected to a database which will be responsible in providing the information based symptoms. On an availability of an internet connection an update will be provided. Main Features: Virtual Doctor is meant for checking up a patient remotely. Symptoms Analyser Emergency Caller Medical library Emergency Contact Register Google Maps to locate nearby hospitals Implementation: J2EE:Java 2 Enterprise Edition: A programming platform which is a part of java platform for developing and running distributed java. Android SDK: SDK stands for Software Development Kit includes a comprehensive set of development tools. These include a debugger, libraries, and a handset emulator. Java JDK: Lays the foundation for the Android SDK. Eclipse Android ADT:Android Development Tools creates the files and structure required for an Android app. SQLite:My SQLite is the world's second most widely used open-source relational database management system (RDBMS).The SQLite phrase stands for Structured Query Language. Android 2.3 and above 512 MB RAM Conclusion: With the completion of our project we have fulfilled most of our project objectives. This application will be available for free in the android market or can be downloaded by using the APK file. This is a simple and user-friendly application that will help the user in knowing at treating the disease. He can also find the alternative medicines in cases of emergencies. This also lets the person know the specialized doctors available in the vicinity. This helps the user to save much of his time and energy. Since it has a simple and interactive user interface, it can be used by all.

Autonomous Navigation using Artificial Neural Networks

Team Member Are :
Damian Fernandes
Lloyd Martiz
Amy Mathew

Project Guide: Ms. Sunitha Guruprasad, Assistant Professor, CSE Department Abstract: The two main challenges of neural network are the learning capability, complex and dynamic navigation environment. In particular, the neural network model is specifically designed for our autonomous navigation system, and a series of training samples are developed to train the neural network. We incorporated stereoscopy with the neural network to solve the problem of autonomous navigation. ANN is a computing system made up of a number of simple, highly interconnected processing elements, which process information by their dynamic state response to external inputs. Stereopsis is the impression of depth that is perceived when a scene is viewed by someone with two eyes and normal binocular vision. In our project we have replaced the human eyes with two cameras each obtaining a slightly different image. We use these images and with the help of VideoIP class and LaneDetect class that holds the algorithms we try to merge these images to a single image and perceive the depth. Main Features: • Stereoscopy formulas are used to obtain the distance and angle. • Neural network is trained using a prerequisite set of values that are extracted dynamically from the environment. The input pattern is tested in the neural network to obtain a single output, i.e. the steering angle.

Connected TV

Team Member Are :
Shashi Kumara H

Project Guide: Ms. Supriya Salian, Assistant Professor, CSE Department Abstract: The Connected TV project is a client server application that allows the user to view/play the files present in his/her mobile devices, like smartphone, tablet etc. on the TV, while still being able to control the TV just as the mobile device is controlled. The mobile devices act as IP remotes to the TV that plays the files. The client receives the input from the user using its user interface. These are sent to the server. The server receives these inputs and performs actions on the file played on the TV. The control information is to be interpreted at real time for better efficiency. The server waits for the client request. When a request arrives, it acknowledges the request by performing the necessary actions. The server is responsible for playing/ displaying the file on the TV it is connected to. Main features: • Removes inconvenience caused by a mobile device due to its screen size. • Convenient in situations where the files are to be sent from multiple sources and also the display is to be controlled by respective sources. • Allows dynamic connections to large number of mobile devices without having to reconnect. • Use of Raspberry Pi as Server

Optical Character Recognition using Artificial Neural Networks (OCRA)

Team Member Are :
Chandini Carolina D’Souza
Jeevan Elroy Martis
Jude Isaac Lobo
Neel Dick Crasta

Project Guide: Dr. Demian Antony D’Mello, Professor, CSE Department Abstract: This project demonstrates how the use of an artificial neural network (ANN) simplifies development of an optical character recognition application, while achieving highest quality of recognition and good performance.OCR deals in recognition and classification of characters from an image. For the recognition to be accurate, certain topological and geometrical properties such as strokes, curves, protrusions, enclosures etc. are calculated, based on which a character is classified and recognized. The feature extraction step of optical character recognition is the most important. A poorly chosen set of features will yield poor classification rates by any neural net-work. Main/Unique features: • The application will be able to recognize any character independent to their size, position, orientation, etc. • Optimize pattern matching algorithm to ensure higher percentage of recognition. • Contains an inbuilt dictionary to spell check and validate words recognized irrespective of whether they are uppercase or lowercase.

Development of a Recursive Navigational Module for Arduino Robot using Artificial Neural Networks

Team Member Are :
Nakshathra Prabhu M
Sharanya Marie Lobo
Shruthi K

Project Guide: Dr Rio D’Souza, Professor, Department of Computer Science & Engineering Abstract: We have developed a recursive navigational module for Arduino robot, making use of Artificial Neural Networks as the core to solve the recursive problem. To reach a reasonable degree of autonomy two basic requirements are sensing and reasoning. The former is provided by onboard sensory system that gathers information about robot with respect to the surrounding scene. The latter is accomplished by devising algorithms that exploit this information in order to generate appropriate commands. We develop a Bidirectional Associative Memory (BAM) which can map between the elements in the problem. Our approach can yield a global navigation algorithm which can be applied to various types of range sensors and mobile robot platforms. Main Features: • Use of Artificial Neural Networks as the core to solve the recursive problem. • Bidirectional Associative Memory (BAM) which maps between the elements in the problem. • Self-built Arduino Robot. • Ultrasonic Sensors for navigation.